Introducing the Matrix M-Stage HPA-3U and HPA-3B, review to follow!

Discussion in 'Headphone Amps (full-size)' started by keanex, Jun 5, 2015.

1. Hi again,

I just found https://cdn.head-fi.org/a/7514507.jpg and looks like the 6K resistor connected to pin-2 of the opamp gets to capacitor's + (plus) and the resistor coming from pin-6 gets to - (minus) of the other capacitor. I wonder if this is not the Ground? 'Cause if it's the ground there, then based on http://www.ti.com/lit/ds/symlink/lme49860.pdf we should have 2 resistors there...though their values are different, but that could be Matrix's choice, of course.

I believe these 2 resistors are the ones taking care of amplifier's gain (along with R7 and R9 I assume, but can't be sure without having the schematic); more details could also be found here: http://www.wakamatsu-net.com/biz/snas435b.pdf (or just read the LME49610 datasheet).

In case someone modified the default internal fixed gain ratio of the amplifier, then all the 3 adjustable gains have been modified too. If I'm not mistaken, the default gain selector for HPA-3U are below:
5dB = 1.75X (3.56V RMS output with 2V input)
10dB = 3.15X (6.13V RMS output with 2V input)
20dB = 10X (20.0V RMS output with 2V input)

If you have another headphone amplifier with a well-known gain, then you could A/B test your HPA-3U against the other amplifier and you could approximate if HPA-3U's gain is the same as Matrix intended.

Also, measuring RMS output voltage with a 1KHz 2V RMS input sine-wave (audiocheck.net perhaps) could give you a good idea about the internal gain of your amplifier. A scope would be perfect for this, but also a true-RMS volt-meter will do to.

Good luck!

P.S.: I'm pretty sure those resistors should not change sound quality of how the HPA-3U sounds, so you could just ignore them and enjoy the sound.

Last edited: May 7, 2017
2. Hi,
I replaced few weeks ago WIMA caps with Audyn Plus MKP (800VDC) and they produced a very positive emotion. Highly recommended upgrade.
I noticed,
- clarity, openess*, transparency and good speed, rhythm and timing compared with WIMA,
- good seperation of the individual instruments with a realistic amount of fine detail, more air,
- sound-stage is spatial with lots of depth! a well controlled and focussed image,
- more noticeable highs!

*Note, It took several houndred hours of playback (minimum 500 hours of burn-in) time before a certain perception of constrained treble energy became notably more open and dynamic sounding and musicality comes back with more natural tone/texture to the sound.

Muses02 Op-amp,
Voltage regulators,
Elna Silmic II* (470uf) instead Nichicon Muse (330uf), 6 pieces
Audyn Plus MKP (800V) instead WIMA capacitors (in the input signal path), 2 pieces
The blue led with a red ones.
*Silmic II are definitely better than Nichicon Muse! (Elna Silmic II capacitors have a nice fleshy low and medium frequencies and a rich harmonic texture. The texture and timbre of acoustic instruments seems very realistic)

Conclusions

WIMA caps are weak there (they are placed in signal path, and the sound is flat, lacking depth and without dynamics!)

Audyn Plus MKP capacitors are good, but just after 600 hours of burning. , they have one of the best price / quality ratio. I paid about 10 euros/pieces for them. Please note that they does not fit into another area becouse are too big for Matrix. I put some pictures to see what I meant.

Last edited: Nov 26, 2017
trybeingarun and raoultrifan like this.
3. Kind of overkill I'd say, but looks quite interesting.

If the longer wires are not adding noise/buzz than it's perfect, just hope your new MUSES will not be suffocated/overheated.

Happy listening!

4. No. Not adding noise/buzz in any case. But the sound is incredible, it seems it is a new DAC/amp. I have the impression that I'm listening to something much more expensive, and that sounds High-End!!

5. Hi guys. Interested in HPA-3U+ version with CS4398 DAC. Is it good enough, higher level than Asus Xonar D1?
I like Vivo Xplay's sound (CS4398+CS8422+OPA2604), so I look forward for this DAC/AMP with same DAC.
I want use it with 42ohm ATH ESW-11 headphones.

Last edited: May 7, 2017
6. Matrix HPA-3U it is comparable with Burson Soloist and Conductor. Asus Xonar D1!? Come on!

7. Here's a thermal image of an HPA-3U in case anybody's interested

8. I have a hpa-3u+, which i've been using for the past 6 months. I also recently purchased Aune X1s & I find Aune X1s to give a much wider soundstage & cleaner output than hpa3u+ (with coin modded hd650). When I use the aune as the dac and matrix as the amp the hpa-3u+'s the setup's performance becomes a little bit better compared to Aune. Now I am a little disappointed with my matrix setup, as it struggles to compete with a cheaper unit . Do you guys use an external dac with the unit? If I have to upgrade 3u+ opamp, what would you all suggest?

9. The built-in DAC it is quite good! If you are able to change all 6 Nichicon Muse capacitors with Elna Silmic II and WIMA with Audyn Plus (like me) and if you have the patience for burn in (about 300/400 hours) you will touch that sound you surely want!
It would be an inexpensive upgrade but worth every penny!!! But then you have to keep it the stock opamp or to take into account the Muses02.
Mine Matrix sounds incredibly good now. The differences are huge, trust me!

trybeingarun likes this.
10. Sure enough, your changes are quite inspiring. I personally have no experience with soldering & have to figure out how I can get it done if I go that route. It's interesting that you find this dac good; I also liked it until I paired it with Aune. It feels like Aune's attack is much better/faster than Matrix's also the soundstage seems a lot wider. It could be that I am interested in that sound just because it's a new unit. When I use Matrix as the dac & Aune as the amp this combination sounds too warm for my liking.
Also, in my Matrix unit dsd sounds very soft. For some reason when I pick dsd (Audirvana, HQ-player doesn't matter) the left & right channels are swapped. I can perfectly live without dsd capability, but a usable one would have been nice.

11. Those WIMA capacitors in signal path are weak. Poor attack (as you noticed), without dynamics, poor texture, narrow sound-stage! The stock Matrix sounds mediocre compared with the big fish IMHO. Unconvincing. I was about to sell it...Now I know I'll keep it for a few years because I can not find anything better in this money. I knew how it sounded Michael Jackson - Bad Album with Stax and eXStatA. I had the impression I was there close to music. I had a source in which I spent a thousand dollars in high-end capacitors (V-Cap CuTf, Black Gates). Now I'm very close with cheap capacitors. Implementation plays an important role, and to find the key points which you can improve, also.

trybeingarun likes this.
12. This makes me really want to try your mods. I will try and find someone who can do these mods for me!

13. Muses02's operating voltage +-16V vs +-22V for the stock. Hope you've taken that into consideration.

14. People change the voltage regulators to lower the voltage.